[conspire] Ah, another test subject for this Saturday's CABAL

[Rick Moen]

Quoting David Fox (dfox94085 at gmail.com):

> Partially, at any rate. So far I am seeding both flavors of ubuntu
> 8.04 (i386/x86_64), and have downloaded the KDE 4.0 edition of Kubuntu
> on i386.

Super.  I assume:  Desktop edition only, not Alternate?  I'll
concentrate on getting the Alternate images of Ubuntu/Kubuntu
i386/x86_64 -- and all the Xubuntu images.  

(Personally, I always install *buntu from Alternate images.  The Desktop
ones I'd use, if at all, only as live CDs -- but I have other live CD
distross I prefer.)

> I haven't gotten the "other" Kubuntu - it seems there are two distinct
> flavors of kubuntu in Hardy. 

Hmm, I'm guessing that means with KDE 3.5.x and with 4.0.  (/me checks.)
Yep.  KDE 4.0 is said to be still pretty much scary-beta stuff, FYI.
People _either_ insanely curious about 4.0's new features _or_ intending
to send in detailed beta-testing bug reports might be interested at this
point.  Others should steer clear.

> This sort of "hosted" stuff - application servers rather than just
> running the stuff locally on your own PC has me worried and somewhat
> skeptical.

Well, you're not alone.  

Let me give you my unguarded personal opinion first, and then I'll see
if I can paint a word-picture of SaaS the way _proponents_ would.

Back in the bad old days, all of our data was locked up in proprietary
formats managed by proprietary software.  Whether that software was
hosted elsewhere or local, either way, you didn't control your data or
the code that managed it.  Locally stored data at least gave you
physical custody -- along with the responsibility for data backup, etc.

Free software / open source fixed the control-of-code problem, and also
the locked-data-formats problem.  Truly autonomous computing without
pain became practical, by my reckoning, around June 2001, when Mozilla
0.9.2 came out, finally replacing the last bit of essential proprietary 
code for general-purpose computing (Netscape Communicator) with
something better.

Looking from my perspective, SaaS is an inherently unattractive retreat
from the gains open source won us.  In one blow, you lose control of
_both_ code _and_ your data.  You don't even have your data locally at
all.  It's in the hands of some company run by strangers whose doings
and priorities, let alone business prospects, you have no clue about.


Proponents, of course, see it differently:  They say that putting your
data in the hands of people who're paid professionals is only prudent,
that they'll take care of your data better than you would, that you can
encrypt it if it's sensitive, that they're bound by contract to protect
your privacy interests (not entirely true), and that users don't _want_ 
the autonomy that I value, because it comes with responsibilities they
can't or don't want to handle.